Drexel University Home Pagewww.drexel.edu DREXEL UNIVERSITY LIBRARIES HOMEPAGE >>
iDEA DREXEL ARCHIVES >>

iDEA: Drexel E-repository and Archives > Drexel Academic Community > College of Information Science and Technology > Faculty Research and Publications (IST) > Case-based reasoning for cash flow forecasting using fuzzy retrieval

Please use this identifier to cite or link to this item: http://hdl.handle.net/1860/2679

Title: Case-based reasoning for cash flow forecasting using fuzzy retrieval
Authors: Weber-Lee, Rosina
Barcia, Ricardo Miranda
Khator, Suresh K.
Issue Date: 1995
Publisher: Springer Verlag
Citation: Case-Based Reasoning Research and Development: Proceedings of the 1st International Conference on Case-Based Reasoning, ICCBR-95: pp. 510-519.
Abstract: Case-Based Reasoning systems use successful past experiences to solve similar problems simulating human approach. Although CBR systems differ from one another, the basic procedure starts with a retrieval method that searches for similar cases in comparison to an input problem. This method must retrieve the case most similar to the input problem resulting the best match. An adaptation phase checks whether the solution of the best match can be readily used to solve the input problem. When the input problem is solved, the adapted solution can be added to the memory of cases. The nature of the cases and their retrieval methods may vary. Section 4 describes Fuzzy Set Theory and how its mechanisms are used to improve the evaluation of similarities on retrieval. The present system uses a fuzzy retrieval and it aims to forecast accounts in cash flows. The Case-Based Reasoner is implemented under an Intelligent Hybrid System (IHS) in a unit called Case-Based Forecasting Unit (CBFU). Before detailing the unit, let us briefly describe its environment (Weber-Lee et al. 1995). The IHS consists of units performing different functions. The project of IHS is object-oriented although some units are still in prototype phase and their modeling into the object-oriented paradigm is being studied. Some of the units of the Intelligent Hybrid System are: Firm, the unit that embodies every function of the company; Case-Based Forecasting Unit, the CBFU is discussed below separately; Database, it keeps the operations and feeds CBFU with all actual values; Interface, there is the interface to input operations and the interface of decision support; Expert System, the expert system unit is the one that manages the IHS’s interface and asks cash flow forecasts for CBFU. The Expert System1 (ES) designed for IHS manages the interface with the user and its knowledge representation is also object-oriented. Hence, the CBFU is a CBR application that forecasts cash flows in the IHS that is the Working Capital decision support system. This paper proposes a fuzzy retrieval to improve the CBR application. Next section presents the importance of cash flow forecasting and how it is linked to the WC management. Then, the following section shows the 1 See Expert Systems: Principles and Programming, (Giarratano, 1994). advantages on the use of CBR as the forecasting technique. Section 3 describes the CBFU unit and its components. Lastly, section 4 presents the fuzzy retrieval, how it is implemented and the meaning of some of the Fuzzy Set Theory concepts applied.
URI: http://hdl.handle.net/1860/2679
Appears in Collections:Faculty Research and Publications (IST)

Files in This Item:

File Description SizeFormat
2006175393.pdf86.61 kBAdobe PDFView/Open
View Statistics

Items in iDEA are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! iDEA Software Copyright © 2002-2010  Duraspace - Feedback